Search

Air pollution prevention and control is an important way to eliminate haze, and air quality prediction can provide predictive information for air pollution prevention and people's health. Therefore, it is of great practical usefulness to establish a scientific and effective air quality prediction model. Combined forecasting is a popular statistical method for air pollution forecasting, which generally includes two steps of individual model selection and weight determination. This research introduces LASSO for individual models selection, and ELM to establish a nonlinear combined forecasting model, named CEEMD-LASSO-ELM. LASSO can not only select individual models, but also avoids the possible collinearity between the selected individual models. Daily air quality index (AQI) series of Guangzhou, Kunming, Lanzhou and Hulunbuir are selected to verify the feasibility and effectiveness of the proposed CEEMD-LASSO-ELM model. Combined with CEEMD, seven comparative models are Random-ELM, Rank-ELM, Random-LR, Rank-LR, LASSO-GRNN, LASSO-PSOGSASVR and LASSO-LR. The data analysis shows that the proposed CEEMD-LASSO-ELM model has better prediction accuracy and stronger generalization ability. Taking Lanzhou as an example, the MAPE of CEEMD-LASSO-ELM model is 1.813% lower than that of the seven comparative models on average. [ABSTRACT FROM AUTHOR]